Integrating Social Networks and Remote Patient

198

MEDINFO 2017: Precision Healthcare through Informatics A.V. Gundlapalli et al. (Eds.) © 2017 International Medical Informatics Association (IMIA) and IOS Press. This article is published online with Open Access by IOS Press and distributed under the terms of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0). doi:10.3233/978-1-61499-830-3-198

Integrating Social Networks and Remote Patient Monitoring Systems to Disseminate Notifications Hugo A. Ribeiro, Eliseu Germano, Sergio T. Carvalho, Eduardo S. Albuquerque Instituto de Informática, Universidade Federal de Goiás, Goiânia, Goiás, Brazil

Abstract Healthcare workforce shortage can be compensated by using information and communication technologies. Remote patient monitoring systems allow us to identify and communicate complications and anomalies. Integrating social networking services into remote patient monitoring systems enables users to manage their relationships. User defined relationships may be used to disseminate healthcare related notifications. Hence this integration leads to quicker interventions and may reduce hospital readmission rate. As a proof of concept, a module was integrated to a remote patient monitoring platform. A mobile application to manage relationships and receive notifications was also developed. Keywords: Social Networking; Patient Monitoring; Physiology

Methods Our approach to enable notification dissemination in a remote patient monitoring system consists of the integration of social network services. This research evolved over three phases leading to a proof of concept: conceptual phase, design phase, and application development. Conceptual Phase Remote patient monitoring systems are specializations of context-aware systems. In this case, context refers to the patients’ health status. A context-aware system architecture, as proposed by [13, 14] is shown in Figure 1.

Introduction Healthcare systems might not properly accommodate all patients in the near future. There is a shortage of healthcare professionals according to World Health Organization [1]: healthcare workforce will be short by 12.9 million professionals in 2035. Also, the world population is demanding more care given the increased life expectancy and the occurrence of non-communicable diseases [2]. Remote patient monitoring systems may reduce the stress on healthcare systems. The stress relief is achieved via the use of technologies that enable communication between patients and health professionals at any time. Using information and communication technologies might also reduce personalized care delivery costs for patients who need to constantly visit a healthcare facility. There is evidence that 33% of patients, especially elderly individuals, want to have access to remote patient monitoring technologies. Furthermore, 40% of those elderly patients want technologies that notify their caregivers whenever there is an emergency situation [3]. Social network services can be integrated into remote patient monitoring systems [4-12]. This integration enables users to manage their relationships. User-defined relationships are used to disseminate notifications. In emergency situations, notifications lead to quicker interventions and may reduce hospital admission rate. This paper presents an architecture for notification emission based on integrating social network services to a remote patient monitoring system. As a proof of concept a module was integrated into a remote patient monitoring platform. A mobile application to manage relationships and receive notifications was also developed.

Figure 1 – Context-aware systems architecture, adapted from [13,14]

Context is acquired (Acquisition Layer) using sensors (Sensing Layer); then represented (Preprocessing Layer), persisted and made available (Storage and Management Layer) to the applications residing on the top layer. Social networks model and manage relationship contextual information [15]. As a context-aware system, a remote patient monitoring system may use the relationship between entities to provide contextual information [16]. Design Phase Following the pattern present in social network systems, such as Facebook and Twitter, a user receives notifications regarding his/her relationships. A relationship management service can be seen as an event notification service in a publish/subscribe architecture [17]. Creation and destruction of relationships are mapped to subscribe and unsubscribe operations as shown in Table 1. Subscriptions are content-based [18]. Messages sent to the event notification service have a field specifying the user who subscribed to the service. Information in this field is used to forward notifications.

H.A. Ribeiro et al. / Integrating Social Networks and Remote Patient Monitoring Systems to Disseminate Notifications

Table 1 – Relationship manager correspondence with event notification service

Operation

199

Results Architecture

Relationship manager Create relationship

Subscribe

Destroy relationship

Unsubscribe

Event notification

Figure 2 shows a message example formatted using JSON.

Figure 2 – Message formatted using JSON

Remote patient monitoring systems following this approach are in accordance with the steps proposed by Morrisey [3]; particularly when it comes to notifying designated responders, initiating rapid interventions.

Our proof of concept aims to deliver notifications to the caregivers. Notifications are sent whenever the collected physiological data presents an anomaly. Patients’ relationships define the notification recipients. Figure 3 outlines the architecture. Physiological data are collected and transmitted to the UbiCare platform (1, 2 and 3). Users can also interact with the platform using other applications that may be developed (3). When a user accesses the UbiCare Social application, a device identifier is requested (A) and set (B). This device identifier is maintained in the UbiCare platform (C). To distribute a notification, the UbiCare platform sends a message and a device identifier list to the push service (D). Device identifiers are processed by the push service and the messages are sent to the caregiver and family devices.

Scenario

Figure 3 – Architecture Application Development Two applications were developed: a relationship manager and a mobile application. The Relationship Manager is a module integrated into UbiCare, a remote patient monitoring system being developed at Instituto de Informática [19]. This module was built using Flask 1, a web development framework for Python. Relationship data are stored in a MySQL database. The mobile application, UbiCare Social, consumes the services provided by the Relationship Manager, allowing users to manage their relationships and to receive notifications. This application runs on Android operating system. In addition, Google Cloud Messaging (GCM) push service was used to distribute notifications.

Consider a scenario with three actors: one patient, one healthcare professional and the patient’s relative. Their relationships are configured as shown in Figure 4.

Figure 4 – Relationship instance

In a situation in which the patient is an elderly person in postoperative care, remote patient monitoring systems allow the patient to be transferred to home, reducing the number of 1

http://flask.pocoo.org/

200


inpatients. Healthcare professionals and family members can assist the patient at home, with minimum impact on the pateint’s privacy, as they will be notified whenever a physiological anomaly is detected.

Dedicated social networks, such as the one presented in this paper, have the potential to mitigate data ownership, functionalities extension, and trust issues; especially in a research scenario.

UbiCare Social Relationship Manager services are accessed via an Android application, named UbiCare Social. Considering the Healthcare Professional from Figure 4, UbiCare Social behaviour is as shown in Figures 5(a), 5(b), 5(c) and 5(d). Whenever an anomaly is detected, a message with the Patient identifier is sent to the Event Notifier. Carers' device identifiers are retrieved and the push service (GCM) is triggered. Healthcare Professional and Family Member will receive a notification as the one show in Figure 6.

Discussion There are research efforts to distribute healthcare related notifications using social network systems [4-12]. Facebook, Twitter, and Google Plus were studied in order to check their suitability to solve this problem [6, 8]. Popular social network platforms have big user bases and provide services via APIs. Despite this, users have minimum control over their data and the platform functionalities. Furthermore, users tend to lack confidence in popular social network platforms.

Figure 6 – Notification issued by the UbiCare platform

Conclusion Social network services may enhance remote patient monitoring systems. This integration leads to quicker interventions and may reduce hospital readmission rate. Further research can be undertaken in the following areas: user acceptance testing, security, privacy, conformance with norms and standards, inclusion of social media services such as an audio-visual communication channel.

Acknowledgements The research for this paper was financially supported by CAPES (Coordination for the Improvement of Higher Education Personnel).

References (a) Login screen

(c) Users interested in the Healthcare Professional

(b) Healthcare professional interests

(d) Other users

Figure 5 – Mobile application behaviour

[1] World Health Organization, Global health workforce shortage to reach 12.9 million in coming decades, WHO (2013), Available at: http://www.who.int/mediacentre/news/releases/2013/healthworkforceshortage/en/ Last access: 12/19/2016. [2] World Health Organization, World report on ageing and health, WHO (2013). [3] J. Morrissey, Remote patient monitoring: How mobile devices will curb chronic conditions: Factor in the time and cost of implementing datadriven technology, Medical Economics (2014), Available at: http://medicaleconomics.modernmedicine.com/medicaleconomics/content/tags/chronic-care/remote-patient-monitoring-howmobile-devices-will-curb-c?page=full Last access: 12/19/2016. [4] A. Bhowmik, F. Khendek, M. Hormati, and R. Glitho, An architecture for m2m enabled social networks, In 14th Annual Mediterranean Ad Hoc Networking Workshop (2015), 1–8. [5] A. Triantafyllidis, V. Koutkias, I. Chouvarda, and N. Maglaveras, A pervasive health system integrating patient monitoring, status logging, and social sharing, IEEE Journal of Biomedical and Health Informatics (2013), 17(1):30–37. [6] C. Khorakhun, and S. Bhatti, Alerts for remote health monitoring using online social media platforms, In IEEE 15th International Conference on e-Health Networking, Applications Services (2013), 177–181. [7] C. Khorakhun, and S. Bhatti, Remote health monitoring using online social media systems, In 6th Joint IFIP Wireless and Mobile Networking Conference (2013), pages 1–8. [8] C. Khorakhun, and S. Bhatti, Using online social media platforms for ubiquitous, personal health monitoring, In IEEE 16th International Con-


[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17] [18]

[19]

ference on e-Health Networking, Applications and Services (2014), 287– 292. C. Khorakhun, and S. Bhatti, Wellbeing as a proxy for a mhealth study, In IEEE International Conference on Bioinformatics and Biomedicine (2014), 32–39. C. Norval, J. Arnott, N. Hine, and V. Hanson, Purposeful social media as support platform: Communication frameworks for older adults requiring care, In 5th International Conference on Pervasive Computing Technologies for Healthcare (2011), 492–494. M. Domingo, A context-aware service architecture for the integration of body sensor networks and social networks through the IP multimedia subsystem, IEEE Communications Magazine (2011), 49(1):102–108. Y. B. D. Trinugroho, F. Reichert, and R. Fensli, eHealth Smart Home Environment Service Platform - Enabling Remote Monitoring and Service Composition through Social Media, In Proceedings of 5th International Conference on Health Informatics (2012), 434–438. M. Baldauf, S. Dustdar, and F. Rosenberg, A survey on contextaware systems, International Journal of Ad Hoc and Ubiquitous Computing (2007), 2(4):263–277. G. W. Musumba, and H. O. Nyongesa, Context awareness in mobile computing: A review, International Journal of Machine Learning and Applications (2013), 2(1). A. M. Kaplan, and M. Haenlein, Users of the world, unite! The challenges and opportunities of social media, Business Horizons (2010), 53(1):59–68. A. Zimmermann, A. Lorenz, and R. Oppermann, An operational definition of context, In Modeling and using context, Springer (2007), 558– 571. V. J. Setty, Publish/Subscribe for Large-Scale Social Interaction: Design, Analysis and Resource Provisioning, PhD Thesis, 2015. P. T. Eugster, P. A. Felber, R. Guerraoui, and A.-M. Kermarrec, The many faces of publish/subscribe. ACM Computing Surveys (2003), 35(2):114–131. E. Germano, D. Battisti, H. A. Ribeiro, and S. T. Carvalho, Ubiquitous Care Plan for Monitoring Patients at Home (in Portuguese). In Congresso Brasileiro de Informática em Saúde (2016), 849-858.

Address for correspondence Hugo A. Ribeiro [email protected]

201